Distributed Clustering in Cognitive Radio Ad Hoc Networks Using Soft-Constraint Affinity Propagation

Abstract

Absence of network infrastructure and heterogeneous spectrum availability in cognitive radio ad hoc networks (CRAHNs) necessitate the self-organization of cognitive radio users (CRs) for efficient spectrum coordination. The cluster-based structure is known to be effective in both guaranteeing system performance and reducing communication overhead in variable network environment. In this paper, we propose a distributed clustering algorithm based on soft-constraint affinity propagation message passing model (DCSCAP). Without dependence on predefined common control channel (CCC), DCSCAP relies on the distributed message passing among CRs through their available channels, making the algorithm applicable for large scale networks. Different from original soft-constraint affinity propagation algorithm, the maximal iterations of message passing is controlled to a relatively small number to accommodate to the dynamic environment of CRAHNs. Based on the accumulated evidence for clustering from the message passing process, clusters are formed with the objective of grouping the CRs with similar spectrum availability into smaller number of clusters while guaranteeing at least one CCC in each cluster. Extensive simulation results demonstrate the preference of DCSCAP compared with existing algorithms in both efficiency and robustness of the clusters.